Image edge extraction algorithm based on adaptive Wiener filtering. (27th May 2022)

• Record Type: Journal Article
• Authors: Sun, Jialong; Liu, Jinlei; Zhang, Zhengyang; Qin, Jiangtao; Yan, Yonghao; Wang, Lize
• Is Part Of: International journal of information and communication technology. Volume 20:Number 4(2022)
• Page Start: 391
• Page End: 410
• Publication Date: 2022-05-27
• Subjects: classical Wiener filter -- adaptive Wiener filter -- Gaussian noise -- noise variance -- edge detection; Information technology -- Periodicals; Computer science -- Periodicals; Telecommunication -- Periodicals; 004.05
• Languages: English
• ISSNs: 1466-6642

Abstract

In this paper, we propose an improved edge extraction algorithm based on adaptive Wiener filtering, aiming at the disadvantage of extracting easily lost edge information from the edge of Canny operator. In this method, the image processed by Wiener filtering is convoluted. Whereas, the classical Wiener filter cannot meet the requirements of noising. Accordingly, a new adaptive Wiener filter is proposed. To begin with, the noise variance of the whole frame image is estimated by a new method, then the image is denoised with four templates. Finally, according to the average value of the difference between all pixel points before and after processing, we can judge which template is used to process the best result. Then the results are applied to the new edge extraction algorithm. The simulation results show that this method effectively preserves the edge information of image processing and suppresses generation of the edge.